def __init__(self, dirpath, az_prjname_by_subprj, samplesheet=None):
info('Parsing the NextSeq500 project structure')
self.kind = 'nextseq500'
DatasetStructure.__init__(self,
dirpath,
az_prjname_by_subprj,
samplesheet=samplesheet)
info('az_prjname_by_subprj: ' + str(az_prjname_by_subprj))
verify_dir(self.unaligned_dirpath, is_critical=True)
for pname, project in self.project_by_name.items():
az_proj_name = az_prjname_by_subprj.get(pname) if not isinstance(
az_prjname_by_subprj, basestring) else az_prjname_by_subprj
if az_proj_name is None:
if len(self.project_by_name) > 1:
warn(
'Warn: cannot correspond subproject ' + pname +
' and project names and JIRA cases. '
'Please, follow the SOP for multiple-project run: http://wiki.rd.astrazeneca.net/display/NG/SOP+-+Pre+Processing+QC+Reporting'
)
continue
az_proj_name = az_prjname_by_subprj.values()[0]
project.set_dirpath(self.unaligned_dirpath, az_proj_name)
for sample in project.sample_by_name.values():
sample.source_fastq_dirpath = project.dirpath
sample.set_up_out_dirs(project.fastq_dirpath,
project.fastqc_dirpath,
project.downsample_targqc_dirpath)
self.basecall_stat_html_reports = self.__get_basecall_stats_reports()
self.get_fastq_regexp_fn = get_nextseq500_regexp
def set_dirpath(self, dirpath, az_project_name):
self.dirpath = dirpath
self.az_project_name = az_project_name
verify_dir(self.dirpath, is_critical=True)
merged_dirpath = join(self.dirpath, 'merged')
if verify_dir(merged_dirpath, silent=True):
self.mergred_dir_found = True
self.fastq_dirpath = self.fastqc_dirpath = merged_dirpath
else:
self.mergred_dir_found = False
self.fastq_dirpath = join(self.dirpath, 'fastq')
self.fastqc_dirpath = join(self.fastq_dirpath, 'FastQC')
info()
self.comb_fastqc_fpath = join(self.fastqc_dirpath, 'FastQC.html')
self.downsample_targqc_report_fpath = None
self.project_report_html_fpath = None
self.downsample_metamapping_dirpath = join(self.dirpath,
'Downsample_MetaMapping')
self.downsample_targqc_dirpath = join(self.dirpath,
'Downsample_TargQC')
self.downsample_targqc_report_fpath = join(
self.downsample_targqc_dirpath, 'targQC.html')
self.project_report_html_fpath = join(self.dirpath,
az_project_name + '.html')
def proc_args(argv):
info(' '.join(sys.argv))
info()
description = 'This script generates target QC reports for each BAM provided as an input. ' \
'Usage: ' + basename(__file__) + ' sample2bam.tsv --bed target.bed --contols sample1:sample2 -o results_dir'
parser = OptionParser(description=description, usage=description)
add_cnf_t_reuse_prjname_donemarker_workdir_genome_debug(parser)
parser.add_option('-o', dest='output_dir', metavar='DIR', default=join(os.getcwd(), 'seq2c'))
parser.add_option('--bed', dest='bed', help='BED file to run Seq2C analysis')
parser.add_option('-c', '--controls', dest='controls', help='Optional control sample names for Seq2C. For multiple controls, separate them using :')
parser.add_option('--seq2c-opts', dest='seq2c_opts', help='Options for the final lr2gene.pl script.')
parser.add_option('--no-prep-bed', dest='prep_bed', help=SUPPRESS_HELP, action='store_false', default=True)
(opts, args) = parser.parse_args()
logger.is_debug = opts.debug
if len(args) == 0:
parser.print_usage()
sys.exit(1)
if len(args) == 1 and not args[0].endswith('.bam'):
sample_names, bam_fpaths = read_samples(verify_file(args[0], is_critical=True, description='Input sample2bam.tsv'))
bam_by_sample = OrderedDict()
for s, b in zip(sample_names, bam_fpaths):
bam_by_sample[s] = b
else:
bam_by_sample = find_bams(args)
run_cnf = determine_run_cnf(opts, is_wgs=not opts.__dict__.get('bed'))
cnf = Config(opts.__dict__, determine_sys_cnf(opts), run_cnf)
check_genome_resources(cnf)
cnf.output_dir = adjust_path(cnf.output_dir)
verify_dir(dirname(cnf.output_dir), is_critical=True)
safe_mkdir(cnf.output_dir)
if not cnf.project_name:
cnf.project_name = basename(cnf.output_dir)
info('Project name: ' + cnf.project_name)
cnf.proc_name = 'Seq2C'
set_up_dirs(cnf)
samples = [
source.TargQC_Sample(name=s_name, dirpath=join(cnf.output_dir, s_name), bam=bam_fpath)
for s_name, bam_fpath in bam_by_sample.items()]
info('Samples: ')
for s in samples:
info(' ' + s.name)
samples.sort(key=lambda _s: _s.key_to_sort())
target_bed = verify_bed(cnf.bed, is_critical=True) if cnf.bed else None
if not cnf.only_summary:
cnf.qsub_runner = adjust_system_path(cnf.qsub_runner)
if not cnf.qsub_runner: critical('Error: qsub-runner is not provided is sys-config.')
verify_file(cnf.qsub_runner, is_critical=True)
return cnf, samples, target_bed, cnf.output_dir
def __init__(self, dirpath, az_prjname_by_subprj=None, samplesheet=None):
info('Parsing the HiSeq project structure')
self.kind = 'hiseq'
DatasetStructure.__init__(self,
dirpath,
az_prjname_by_subprj,
samplesheet=samplesheet)
verify_dir(self.unaligned_dirpath, is_critical=True)
self.basecall_stat_html_reports = self.__get_basecall_stats_reports()
for pname, project in self.project_by_name.items():
proj_dirpath = join(
self.unaligned_dirpath, 'Project_' + pname.replace(
' ', '-')) #.replace('-', '_').replace('.', '_'))
az_proj_name = az_prjname_by_subprj.get(pname) if not isinstance(
az_prjname_by_subprj, basestring) else az_prjname_by_subprj
if az_proj_name is None:
if len(self.project_by_name) > 1:
warn(
'Warn: cannot correspond subproject ' + pname +
' and project names and JIRA cases. '
'Please, follow the SOP for multiple-project run: http://wiki.rd.astrazeneca.net/display/NG/SOP+-+Pre+Processing+QC+Reporting'
)
continue
az_proj_name = az_prjname_by_subprj.values()[0]
project.set_dirpath(proj_dirpath, az_proj_name)
for sname, sample in project.sample_by_name.items():
sample.source_fastq_dirpath = join(
project.dirpath, 'Sample_' + sname.replace(
' ', '-')) #.replace('-', '_').replace('.', '_'))
sample.set_up_out_dirs(project.fastq_dirpath,
project.fastqc_dirpath,
project.downsample_targqc_dirpath)
basecalls_symlink = join(project.dirpath, 'BaseCallsReports')
if not exists(basecalls_symlink):
info('Creating BaseCalls symlink ' + self.basecalls_dirpath +
' -> ' + basecalls_symlink)
try:
os.symlink(self.basecalls_dirpath, basecalls_symlink)
except OSError:
err('Cannot create symlink')
traceback.print_exc()
else:
info('Created')
if exists(basecalls_symlink):
self.basecalls_dirpath = basecalls_symlink
self.get_fastq_regexp_fn = get_hiseq_regexp
def main():
info(' '.join(sys.argv))
info()
description = 'This script runs preprocessing.'
parser = OptionParser(description=description)
parser.add_option('-1', dest='left_reads_fpath', help='Left reads fpath')
parser.add_option('-2', dest='right_reads_fpath', help='Right reads fpath')
parser.add_option('--sample', dest='sample_name', help='Sample name')
parser.add_option('-o', dest='output_dir', help='Output directory path')
parser.add_option(
'--downsample-to',
dest='downsample_to',
default=None,
type='int',
help=
'Downsample reads to avoid excessive processing times with large files. '
'Default is 1 million. Set to 0 to turn off downsampling.')
add_cnf_t_reuse_prjname_donemarker_workdir_genome_debug(parser, threads=1)
(opts, args) = parser.parse_args()
logger.is_debug = opts.debug
cnf = Config(opts.__dict__, determine_sys_cnf(opts),
determine_run_cnf(opts))
left_reads_fpath = verify_file(opts.left_reads_fpath, is_critical=True)
right_reads_fpath = verify_file(opts.right_reads_fpath, is_critical=True)
output_dirpath = adjust_path(
opts.output_dir) if opts.output_dir else critical(
'Please, specify output directory with -o')
verify_dir(dirname(output_dirpath),
description='output_dir',
is_critical=True)
with workdir(cnf):
sample_name = cnf.sample_name
if not sample_name:
sample_name = _get_sample_name(left_reads_fpath, right_reads_fpath)
results_dirpath = run_fastq(cnf,
sample_name,
left_reads_fpath,
right_reads_fpath,
output_dirpath,
downsample_to=cnf.downsample_to)
verify_dir(results_dirpath, is_critical=True)
info()
info('*' * 70)
info('Fastqc results:')
info(' ' + results_dirpath)
def main():
info(' '.join(sys.argv))
info()
description = 'This script runs preprocessing.'
parser = OptionParser(description=description)
parser.add_option('-1', dest='left_reads_fpath', help='Left reads fpath')
parser.add_option('-2', dest='right_reads_fpath', help='Right reads fpath')
parser.add_option('--sample', dest='sample_name', help='Sample name')
parser.add_option('-o', dest='output_dir', help='Output directory path')
parser.add_option('--downsample-to', dest='downsample_to', default=None, type='int',
help='Downsample reads to avoid excessive processing times with large files. '
'Default is 1 million. Set to 0 to turn off downsampling.')
add_cnf_t_reuse_prjname_donemarker_workdir_genome_debug(parser, threads=1)
(opts, args) = parser.parse_args()
if not opts.left_reads_fpath or not opts.right_reads_fpath or not opts.output_dir:
parser.print_usage()
verify_file(opts.left_reads_fpath, is_critical=False)
left_reads_fpath = adjust_path(opts.left_reads_fpath)
verify_file(opts.right_reads_fpath, is_critical=False)
right_reads_fpath = adjust_path(opts.right_reads_fpath)
output_dirpath = adjust_path(opts.output_dir) if opts.output_dir else critical('Please, specify output directory with -o')
verify_dir(dirname(output_dirpath), description='output_dir', is_critical=True)
left_reads_fpath, right_reads_fpath, output_dirpath =\
map(_proc_path, [left_reads_fpath, right_reads_fpath, output_dirpath])
ssh = connect_to_server(server_url='blue.usbod.astrazeneca.net', username='******', password='******')
fastqc_py = get_script_cmdline(None, 'python', 'scripts/pre/fastqc.py')
fastqc_py = fastqc_py.replace(REPORTING_SUITE_PATH_CLARITY, REPORTING_SUITE_PATH_WALTHAM)
fastqc_py = fastqc_py.replace(PYTHON_PATH_CLARITY, PYTHON_PATH_WALTHAM)
cmdl = '{fastqc_py} -1 {left_reads_fpath} -2 {right_reads_fpath} -o {output_dirpath}'
if opts.sample_name:
cmdl += ' --sample {opts.sample_name}'
if opts.downsample_to:
cmdl += ' --downsample-to ' + str(int(opts.downsample_to))
cmdl = cmdl.format(**locals())
cmdl += ' 2>&1'
info(cmdl)
stdin, stdout, stderr = ssh.exec_command(cmdl)
for l in stdout:
err(l, ending='')
info()
ssh.close()
def check_dirs_and_files(cnf, file_keys=list(), dir_keys=list()):
errors = []
def _verify_input_file(_key):
cnf[_key] = adjust_path(cnf[_key])
if not verify_file(cnf[_key], _key):
return False
if 'bam' in _key and not verify_bam(cnf[_key]):
return False
if 'bed' in _key and not verify_bed(cnf[_key]):
return False
return True
for key in file_keys:
if key and key in cnf and cnf[key]:
if not _verify_input_file(key):
errors.append('File ' + cnf[key] +
' is empty or cannot be found')
else:
cnf[key] = adjust_path(cnf[key])
for key in dir_keys:
if key and key in cnf and cnf[key]:
cnf[key] = adjust_path(cnf[key])
if not verify_dir(cnf[key], key):
errors.append('Directory ' + cnf[key] +
' is empty or cannot be found')
else:
cnf[key] = adjust_path(cnf[key])
return errors
def __find_unaligned_dir(self):
unaligned_dirpath = join(self.dirpath, 'Unalign')
if verify_dir(unaligned_dirpath,
description='"Unalign" directory',
silent=True):
unaligned_dirpath = unaligned_dirpath
else:
unaligned_dirpath = None
warn('No unalign directory')
return unaligned_dirpath
def __init__(self, dirpath, az_prjname_by_subprj, samplesheet=None):
self.az_prjname_by_subprj = az_prjname_by_subprj
illumina_project_name = None
if '/Unalign/' in dirpath:
self.dirpath = dirpath.split('/Unalign/')[0]
self.unaligned_dirpath = self.__find_unaligned_dir()
verify_dir(self.unaligned_dirpath,
description='Unalign dir',
is_critical=True)
illumina_project_name = dirpath.split(
'/Unalign/'
)[1] # something like AURA.FFPE.AZ300, in contast with project_name which is something like Bio_123_AURA_FFPE_AZ300
info('Processing sub-project ' + illumina_project_name)
else:
self.dirpath = dirpath
self.unaligned_dirpath = self.__find_unaligned_dir()
self.basecalls_dirpath = join(self.dirpath,
'Data/Intensities/BaseCalls')
verify_dir(self.basecalls_dirpath, is_critical=True)
self.bcl2fastq_dirpath = None
self.source_fastq_dirpath = None
if samplesheet:
self.samplesheet_fpath = samplesheet
else:
self.samplesheet_fpath = self.__find_sample_sheet()
self.project_by_name = self._parse_sample_sheet(self.samplesheet_fpath)
if illumina_project_name: # we want only a specific project
if illumina_project_name not in self.project_by_name:
info()
critical('Err: project ' + illumina_project_name +
' not in the SampleSheet ' + self.samplesheet_fpath)
else:
self.project_by_name = {
illumina_project_name:
self.project_by_name[illumina_project_name]
}
def __init__(self, dirpath, az_prjname_by_subprj, samplesheet=None):
info('Parsing the MiSeq project structure')
self.kind = 'miseq'
DatasetStructure.__init__(self,
dirpath,
az_prjname_by_subprj,
samplesheet=samplesheet)
base_dirpath = self.unaligned_dirpath
if not verify_dir(base_dirpath, silent=True):
base_dirpath = self.basecalls_dirpath
verify_dir(base_dirpath, description='Source fastq dir')
for pname, project in self.project_by_name.items():
proj_dirpath = join(base_dirpath, pname)
if not verify_dir(proj_dirpath, silent=True):
proj_dirpath = base_dirpath
az_proj_name = az_prjname_by_subprj.get(pname) if not isinstance(
az_prjname_by_subprj, basestring) else az_prjname_by_subprj
if az_proj_name is None:
if len(self.project_by_name) > 1:
warn(
'Warn: cannot correspond subproject ' + pname +
' and project names and JIRA cases. '
'Please, follow the SOP for multiple-project run: http://wiki.rd.astrazeneca.net/display/NG/SOP+-+Pre+Processing+QC+Reporting'
)
continue
az_proj_name = az_prjname_by_subprj.values()[0]
project.set_dirpath(proj_dirpath, az_proj_name)
for sample in project.sample_by_name.values():
sample.source_fastq_dirpath = project.dirpath
sample.set_up_out_dirs(project.fastq_dirpath,
project.fastqc_dirpath,
project.downsample_targqc_dirpath)
self.basecall_stat_html_reports = []
self.get_fastq_regexp_fn = get_hiseq4000_miseq_regexp
def proc_opts():
parser = OptionParser(description='')
(opts, args) = parser.parse_args()
logger.is_debug = opts.debug
if len(args) < 1:
critical('First argument should be a root datasets dir')
# if len(args) < 2:
# info('No dataset path specified, assuming it is the current working directory')
# dataset_dirpath = adjust_path(os.getcwd())
# jira_url = args[0]
root_dirpath = verify_dir(args[0], is_critical=True, description='Dataset directory') # /ngs/oncology/datasets/hiseq/150521_D00443_0159_AHK2KTADXX
info(' '.join(sys.argv))
return root_dirpath
def create_jbrowse_symlink(genome, project_name, sample, file_fpath):
jbrowse_data_path, _, _ = set_folders(genome)
jbrowse_dirpath = join(jbrowse_data_path, 'tracks')
jbrowse_project_dirpath = join(jbrowse_dirpath, project_name)
base, ext = splitext_plus(file_fpath)
if ext in ['.tbi', '.bai']:
base, ext2 = splitext_plus(base)
ext = ext2 + ext
sym_link = join(jbrowse_project_dirpath, sample + ext)
if not verify_dir(jbrowse_project_dirpath):
safe_mkdir(jbrowse_project_dirpath)
if isfile(file_fpath) and not isfile(sym_link):
try:
os.symlink(file_fpath, sym_link)
except OSError:
warn(traceback.format_exc())
if isfile(sym_link):
change_permissions(sym_link)
return sym_link
def _run_multisample_qualimap(cnf, output_dir, samples, targqc_full_report):
""" 1. Generates Qualimap2 plots and put into plots_dirpath
2. Adds records to targqc_full_report.plots
"""
plots_dirpath = join(output_dir, 'plots')
if cnf.reuse_intermediate and verify_dir(plots_dirpath) and [
f for f in listdir(plots_dirpath) if not f.startswith('.')
]:
info('Qualimap miltisample plots exist - ' + plots_dirpath +
', reusing...')
else:
# Qualimap2 run for multi-sample plots
if len(
[s.qualimap_html_fpath
for s in samples if s.qualimap_html_fpath]) > 0:
qualimap = get_system_path(cnf,
interpreter_or_name=None,
name='qualimap')
if qualimap is not None and get_qualimap_type(qualimap) == 'full':
qualimap_output_dir = join(cnf.work_dir,
'qualimap_multi_bamqc')
_correct_qualimap_genome_results(cnf, samples)
_correct_qualimap_insert_size_histogram(cnf, samples)
safe_mkdir(qualimap_output_dir)
rows = []
for sample in samples:
if sample.qualimap_html_fpath:
rows += [[sample.name, sample.qualimap_html_fpath]]
data_fpath = write_tsv_rows(
rows,
join(qualimap_output_dir,
'qualimap_results_by_sample.tsv'))
qualimap_plots_dirpath = join(qualimap_output_dir,
'images_multisampleBamQcReport')
cmdline = '{qualimap} multi-bamqc --data {data_fpath} -outdir {qualimap_output_dir}'.format(
**locals())
res = call(cnf,
cmdline,
exit_on_error=False,
return_err_code=True,
env_vars=dict(DISPLAY=None),
output_fpath=qualimap_plots_dirpath,
output_is_dir=True)
if res is None or not verify_dir(qualimap_plots_dirpath):
warn(
'Warning: Qualimap for multi-sample analysis failed to finish. TargQC will not contain plots.'
)
return None
else:
if exists(plots_dirpath):
shutil.rmtree(plots_dirpath)
shutil.move(qualimap_plots_dirpath, plots_dirpath)
else:
warn(
'Warning: Qualimap for multi-sample analysis was not found. TargQC will not contain plots.'
)
return None
targqc_full_report.plots = []
for plot_fpath in listdir(plots_dirpath):
plot_fpath = join(plots_dirpath, plot_fpath)
if verify_file(plot_fpath) and plot_fpath.endswith('.png'):
targqc_full_report.plots.append(relpath(plot_fpath, output_dir))
def main(args):
if len(args) < 2:
sys.exit('Usage ' + __file__ +
' input.tsv bcbio.csv [dir_with_bams] [bina_dir]')
inp_fpath = args[0]
verify_file(args[0], is_critical=True)
out_fpath = args[1]
verify_dir(dirname(adjust_path(out_fpath)), is_critical=True)
bam_dirpath = None
if len(args) > 2:
bam_dirpath = args[2]
verify_dir(adjust_path(bam_dirpath), is_critical=True)
# bam_opt = args[2]
# try:
# bam_col = int(bam_opt)
# bam_dirpath = None
# except ValueError:
# bam_col = None
# verify_dir(bam_opt, is_critical=True)
# bam_dirpath = args[2]
bina_dirpath = None
if len(args) > 3:
bina_dirpath = args[3]
verify_dir(dirname(adjust_path(bina_dirpath)), is_critical=True)
# filtered_bams_dirpath = adjust_path(sys.argv[3])
# verify_dir(join(filtered_bams_dirpath, os.pardir), is_critical=True)
columns_names = 'study barcode disease disease_name sample_type sample_type_name analyte_type library_type center center_name platform platform_name assembly filename files_size checksum analysis_id aliquot_id participant_id sample_id tss_id sample_accession published uploaded modified state reason'
samples_by_patient = defaultdict(list)
delim = '\t'
barcode_col = 1
bam_col = 13
is_tcga_tsv = True
with open(inp_fpath) as fh:
for i, l in enumerate(fh):
if not l.strip():
continue
if i == 0:
if len(l.split('\t')) == 27:
err('Interpreting as TCGA tsv')
if l.split('\t')[0] != 'TCGA': continue # skipping header
else:
delim = None
for j, f in enumerate(l.split()):
if f.startswith('TCGA'):
barcode_col = j
err('barcode col is ' + str(j))
if f.endswith('bam'):
bam_col = j
err('bam col is ' + str(j))
is_tcga_tsv = False
fs = l.split(delim)
barcode = fs[barcode_col].split(
'-') # TCGA-05-4244-01A-01D-1105-08
sample = Sample()
sample.bam = fs[bam_col]
sample.bam_base_name = basename(os.path.splitext(fs[bam_col])[0])
sample.description = fs[barcode_col]
sample.patient = '-'.join(barcode[:3])
if is_tcga_tsv:
sample.reason = fs[26]
sample_type = int(barcode[3][:2])
if sample_type >= 20 or sample_type <= 0:
continue
sample.is_normal = 10 <= sample_type < 20
sample.is_blood = sample_type in [
3, 4, 9, 10
] # https://tcga-data.nci.nih.gov/datareports/codeTablesReport.htm
if any(s.description == sample.description
for s in samples_by_patient[sample.patient]):
prev_sample = next(s
for s in samples_by_patient[sample.patient]
if s.description == sample.description)
# comp reason
# if 'Fileset modified' not in prev_sample.reason and 'Fileset modified' in sample.reason:
# err('Duplicated sample: ' + sample.description + ' Fileset modified not in old ' + prev_sample.name + ' over ' + sample.name)
# pass
# elif 'Fileset modified' in prev_sample.reason and 'Fileset modified' not in sample.reason:
# samples_by_patient[sample.patient].remove(prev_sample)
# samples_by_patient[sample.patient].append(sample)
# err('Duplicated sample: ' + sample.description + ' Fileset modified not in new ' + sample.name + ' over ' + prev_sample.name)
# else:
# comp version
prev_version = get_bam_version(prev_sample.bam_base_name)
version = get_bam_version(sample.bam_base_name)
err('Duplicated sample: ' + sample.description +
' Resolving by version (' + ' over '.join(
map(str,
sorted([prev_version, version])[::-1])) + ')')
if version > prev_version:
samples_by_patient[sample.patient].remove(prev_sample)
samples_by_patient[sample.patient].append(sample)
else:
samples_by_patient[sample.patient].append(sample)
batches = []
final_samples = set()
if bina_dirpath:
safe_mkdir(bina_dirpath)
for patient, patient_samples in samples_by_patient.iteritems():
tumours = [s for s in patient_samples if not s.is_normal]
normals = [s for s in patient_samples if s.is_normal]
main_normal = None
if len(normals) >= 1:
if any(n.is_blood for n in normals):
main_normal = next(n for n in normals if n.is_blood)
else:
main_normal = normals[0]
if tumours:
for n in normals[1:]:
b = Batch(n.description + '-batch')
b.tumour = n
batches.append(b)
for t in tumours:
b = Batch(t.description + '-batch')
b.tumour = t
t.batches.add(b)
final_samples.add(t)
if main_normal:
b.normal = main_normal
main_normal.batches.add(b)
final_samples.add(main_normal)
batches.append(b)
##################
###### Bina ######
if bina_dirpath:
bina_patient_dirpath = join(bina_dirpath, patient)
safe_mkdir(bina_patient_dirpath)
normals_csv_fpath = join(bina_patient_dirpath, 'normals.csv')
tumours_csv_fpath = join(bina_patient_dirpath, 'tumors.csv')
if main_normal:
with open(normals_csv_fpath, 'w') as f:
f.write('name,bam\n')
bam_fpath = join(
bam_dirpath,
main_normal.bam) if bam_dirpath else main_normal.bam
f.write(main_normal.description + ',' + bam_fpath + '\n')
with open(tumours_csv_fpath, 'w') as f:
f.write('name,bam\n')
for t in tumours:
bam_fpath = join(bam_dirpath,
t.bam) if bam_dirpath else t.bam
f.write(t.description + ',' + bam_fpath + '\n')
if bina_dirpath:
err('Saved bina CSVs to ' + bina_dirpath)
###########################
######## Bcbio CSV ########
print 'bcbio_nextgen.py -w template bcbio.yaml', out_fpath,
with open(out_fpath, 'w') as out:
out.write('sample,description,batch,phenotype\n')
for s in sorted(final_samples, key=lambda s: s.bam_base_name):
out.write(','.join([
s.bam_base_name, s.description, ';'.join(
sorted(b.name for b in s.batches)),
('normal' if s.is_normal else 'tumor')
]) + '\n')
bam_fpath = join(bam_dirpath, s.bam) if bam_dirpath else s.bam
if verify_bam(bam_fpath, is_critical=False):
try:
bam = pysam.Samfile(bam_fpath, "rb")
except ValueError:
err(traceback.format_exc())
err('Cannot read ' + bam_fpath)
err()
# n_rgs = max(1, len(bam.header.get("RG", [])))
else:
print bam_fpath,
def __get_basecall_stats_reports(self):
dirpath = join(self.unaligned_dirpath, 'Reports', 'html')
index_html_fpath = join(dirpath, 'index.html')
if verify_dir(dirpath) and verify_file(index_html_fpath):
return [index_html_fpath]